The Prenatal Determinants of Schizophrenia study will investigate the relation between prenatal exposures and schizophrenia. The study will be based on a cohort of 11,799 individuals born in Oakland in 1959-1966 who have been followed since early gestation as part of the Child Health and Development Study. The cohort are now in the age of risk for schizophrenia. The strengths of the study include a large and representative sample of births, a rich array of data on early prenatal exposures, an excellent registry for ascertainment of cases of schizophrenia and thorough diagnostic assessments of the cases. In addition, stored prenatal maternal sera are available for the collection of new exposure on future studies. The study aims first to investigate whether prenatal exposures are associated with an increased incidence of schizophrenia and of schizophrenia spectrum disorders. The prenatal exposures that can be examined include a broad range of exposures--infectious, nutritional, toxic, immune and hormonal--that could affect brain development. To illustrate our approach, we focus here on the a priori testing of a small number of prenatal exposures that are plausible risk factors for schizophrenia, namely, maternal use of prescribed amphetamines, maternal use of alcohol and rhesus (Rh) incompatibility. The second aim is to suggest causal pathways that may underlie observed associations between prenatal exposures and the risk of schizophrenia. The pathways will be examined using the comprehensive data on maternal and paternal characteristics, intrapatum events and the condition of the infant at birth that are available for this cohort. These data enable a thorough analysis of potential confounding and mediating variables and of effect modification. The third aim is to prepare the groundwork for a future case-control study that will be 'nested' in this cohort. The case-control study will use the stored prenatal maternal sera to obtain new exposure data, also potentially available for a wide range of exposures. To illustrate the necessary groundwork, we describe the preparatory work for an investigation of low-normal maternal folate as a risk factor for schizophrenia. The present application requests three years of funding to complete the PDS-Coh study and only the groundwork for the case-control study; a separate application for the case-control study will be submitted toward the end of that period.